MATLAB Source Code for Face Recognition Using Diagonal DCT and 2D PCA

This document presents a MATLAB source code implementation for face recognition utilizing diagonal Discrete Cosine Transform (DCT) and Two-Dimensional Principal Component Analysis (2D PCA). The code demonstrates a practical approach where diagonal DCT coefficients are first extracted from facial images to capture significant frequency domain features, followed by 2D PCA for dimensionality reduction and feature optimization. This implementation handles critical aspects including image preprocessing, feature matrix computation, and classification using distance metrics like Euclidean or Manhattan distance. The code structure includes modular functions for DCT coefficient extraction along image diagonals, covariance matrix calculation for 2D PCA, and eigenvalue decomposition for optimal feature projection. Particularly valuable for both academic research and practical applications, this implementation can be integrated into security systems, access control mechanisms, and biometric authentication platforms. The algorithm efficiently reduces computational complexity while maintaining recognition accuracy by processing 2D image matrices directly without vectorization. Researchers and developers interested in pattern recognition and computer vision will find this implementation insightful for understanding feature extraction techniques and classification methodologies in image-based biometric systems.